Denitrification, Nitrate Reduction, and Oxygen Consumption in Coastal and Estuarine Sediments

Transcription

1 APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Mar. 1982, p /82/3648-6$2./ Vol. 43, No. 3 Denitrification, Nitrate Reduction, and Oxygen Consumption in Coastal and Estuarine Sediments TAKASHI NISHIO, ISAO KOIKE, AND AKIHIKO HATTORI* Ocean Research Institute, University of Tokyo, Nakano ku, Tokyo 164, Japan Received 16 July 1981/Accepted 12 November 1981 Denitrification and consumption of oxygen and nitrate in sediments from Tama Estuary, Odawa Bay, and Tokyo Bay were measured in an experimental sediment-water system. Filtered seawater containing [15N]nitrate flowed continuously over undisturbed sediments, and the concentrations of 2, nitrate, and nitrite in the influent and effluent and of 15N2 in the effluent were monitored. Under steady-state conditions, the rate of nitrate consumption was the same order of magnitude as the rate of oxygen consumption in Tama Estuary sediments, whereas the former rate was one order of magnitude lower than the latter rate in Odawa Bay and Tokyo Bay sediments. Denitrification accounted for 27 to 57% of the nitrate consumption. Active bacterial denitrification in coastal marine sediments that are rich in organic materials has been reported by Koike and Hattori (5), Srensen (1), and Oren and Blackburn (8). In the methods which these workers used, however, some physical and chemical disturbances inevitably occur during the experiments; it is difficult to assess the extent of the effects of these disturbances on in situ denitrification. We (7a) have demonstrated that the amount of excess N2, as estimated from the N2/Ar ratio in the pore water of an estuarine sediment, is far smaller than the amount expected based on the denitrification rates determined by bottle incubation experiments. We concluded that bacterial denitrification proceeds actively but that excess N2 produced by denitrification is lost rapidly to the overlying water because of physical and biological disturbances of the near-surface sediment. In this study, we attempted to determine denitrification rates in an experimental sediment-water system in which filtered seawater containing [15N]nitrate flowed at a constant rate over undisturbed coastal and estuarine sediments. The rates of oxygen and nitrate consumption were determined simultaneously and were compared with the rate of denitrification. MATERIALS AND METHODS Seawater and sediment samples were collected at Tama Estuary, Odawa Bay, and Tokyo Bay (Fig. 1). The sampling dates and water depths are given in Table 1. The sediment samples from Tama Estuary and Odawa Bay were collected with Plexiglas tubes (inner diameter, 11.4 cm) by hand, and those from Tokyo Bay were collected with a gravity core sampler (tubes with an inner diameter of 11 cm) on the KT cruise of RNV Tansei Maru, University of Tokyo. Judging from the color, all of the sediment cores were anoxic except for a thin surface layer. Special care was taken to preserve sediment structure during sampling and transportation. The sediment structures were somewhat distorted near the inner walls of the Plexiglas tubes, but this effect could be disregarded because of the large diameter of the samplers. Overlying water samples were collected with Plexiglas bottles or van Dorn bottles. Other sediment samples were collected with a Plexiglas tube (inner diameter, 3.5 cm). The sediment cores were sectioned into segments between 1 and 3 cm long, and pore waters were extracted by centrifugation at 2, rpm for 1 min and were kept froen until analysis of nitrate and nitrite. The analysis was completed within a few weeks. The experimental setup used is illustrated in Fig. 2. A Plexiglas tube containing a core of sediment 2 to 3 cm long and 5 to 1 cm of overlying water was placed in a water bath and maintained at the in situ temperature. A rubber stopper was fixed 4 to 5 cm above the sediment so as not to include any air bubbles. The Plexiglas tube was wrapped with aluminum foil to keep out light. Experiments were started within 1 h after sampling. Whatman GF/C-filtered seawater to which 1 or 2,uM (final concentration) [15N]nitrate (5 atom% "5N) was added was placed in a reservoir and run over the sediment at a fixed flow rate of 5 to 1 ml/ min. The water layer was mixed gently and continuously with an Acrobat stirrer (MS Instrument, Osaka, Japan). The turnover rate of the overlying water (flow rate divided by the volume of overlying water) was 1 to 2 turnovers per hr. The dissolved 2 concentration in the effluent was monitored continuously with an oxygen electrode (Yellow Springs Instrument Co., Yellow Springs, Ohio). The experiments were continued for 4 to 1 h. At 3- to 6-h intervals, portions of the effluent were collected in vacuum-tight test tubes (volume, 13 ml) for determinations of evolved 15N2. The tubes were capped with rubber stoppers so as not to include any air bubbles, and.2 ml of.1 M HgCl2 was added Downloaded from on January 3, 219 by guest

2 VOL. 43, 1982 DENITRIFICATION IN SEDIMENTS *3' through each stopper with a hypodermic syringe needle to stop biological reactions. Additional subsamples (about 2 ml) were collected simultaneously and froen immediately for later determinations of concentrations of nitrate and nitrite. The 15N content in the dissolved N2 was determined as described by Koike et al. (7), using a Hitachi model RMU-6 mass spectrometer fitted with a double collector and a dual inlet system for ratiometry. Nitrate and nitrite were determined with a Hirama autoanalyer. We used the method of Wood et al. (11) for nitrate and the method of Bendschneider and Robinson (1) for nitrite, with minor modifications. The production and consumption of N2, 2, and nitrate in the overlying water were small and could be neglected, because the majority of the microorganisms in this water had been removed by filtration. Therefore, at steady state, the fluxes (F) of N2, 2, and nitrate across the sediment-water interface could be calculated by the following equation: F = (Ci A, where A is surface - C)V/ area of sediment, V is the flow FIG. 1. Locations of the sampling sites. rate, and Ci and C, are the concentrations in the influent and effluent, respectively. The fluxes of 2, nitrate, and N2 should have been equivalent to the consumption rate of oxygen, the consumption rate of nitrate (plus nitrite), and the production rate of N2 in the sediment, respectively. RESULTS Figure 3 shows the vertical profiles of nitrate plus nitrite in the sedimentary pore waters and overlying waters. The concentrations of nitrate plus nitrite in the overlying waters ranged from 17.4 to 114,LM, but those in the sedimentary pore waters were very low. Figure 4 shows the changes with time in the concentrations of dissolved 2, nitrate, and nitrite during the experiments. The concentrations of dissolved 2 and nitrate changed similarly. The concentrations of nitrite remained approxi- Downloaded from on January 3, 219 by guest TABLE 1. Characteristics of the sampling stations Location Sampling date Depth Temp Overlying water (m) (OC) NO3- concn (>±M) NO2- concn (>.M) Tama Estuary 7 May 198.5, 2.a June 198.5, 2.a Odawa Bay 15 May 198.5, 2.a June 198.5, 2.Oa b Tokyo Bay Station 2 1 September Station 8 1 September a The first value is the depth at low tide, and the second value is the depth at high tide. b NO2 was included.

3 65 NISHIO, KOIKE, AND HATTORI APPL. ENVIRON. MICROBIOL. WATER BATH FIG. 2. Experimental setup used for the simultaneous determination of oxygen consumption, and N2 production in the sediments. mately constant throughout the experiment. Steady levels of dissolved 2 and nitrate in the effluent were achieved within 3 h, and these levels were maintained for 2 to 3 h. A later E N( 3 * N 3( ML) L ( 11, U OVERLYING WATER UOL-=:=?- - e - 51 = 1 1- a.. w a I 151 Ii I + i I consumption, nitrate decline in the dissolved 2 and nitrate concentrations could be attributed at least in part to the growth of microorganisms on the walls of the Plexiglas tubes and connecting tubes and on the _- SEDIMEENT TAMA ESTUARY 7 MAY '8 ODAWA BAY 15 MAY '8 TOKYO BAY (Sta.2) 1 SEPT. '8 I i 12 Downloaded from on January 3, 219 by guest FIG. 3. Vertical profiles of nitrate plus nitrite in sedimentary pore waters of Tokyo Bay, Odawa Bay, and Tama Estuary.

4 VOL. 43, 1982 DENITRIFICATION IN SEDIMENTS ! ~~* A f~~~~~~~~~~~~~~~~~~~~~~~~~~~ -36 _ - Z Oi- IC26. ODAWA BAY 15 MAY '8 *- -'-a. 2' ~~~~~~~ % 7NO O o -. T W- 2 4 O 6 s 1 T I ME (hours) 1 4 Li TAMA ESTUARY 19 JUNE '8 g2 e ^ NO3 o o 2ANO; -ou 1 2 a TOKYO BAY Sta.2 1 SEPT.' SO _261 NO; NO-2 l FIG. 4. Changes with time in concentrations of dissolved 2, nitrate, and nitrite in the experimental sediment-water system. Symbols:, concentrations in influents;, concentrations in effluents. Samples collected from Tokyo Bay (Station 2, 1 September 198), Odawa Bay (15 May 198), and Tama Estuary (7 May 198) were used in these experiments. bottom surfaces of the rubber stoppers. Actually, the 2 and nitrate concentrations in the effluent increased when the inner surfaces of the rubber stoppers were cleaned with gaue and the connecting tubes were replaced with new tubes, although complete recovery to the steady level did not occur. Changes in the oxidation-reduction state of the system during prolonged incubation and removal of soft materials from the surface sediments were probably other reasons for this decline. Figure 5 shows the time courses of N2 evolution from nitrate by denitrification. The amounts of N2 evolved were calculated from the 15N content of the dissolved N2 and nitrate. Initially, the concentration of evolved N2 in the overlying water increased, but after 1 h from the onset of the experiment this concentration remained practically constant. This indicates that most of the denitrification occurred near the surface of the sediments because bacterial nitrate consumption was fast and that the concentration of evolved N2 in the sedimentary pore water near the sediment-water interface reached a steady level within 1 h. Fluxes of 2, nitrate plus nitrite, and N2 across the sediment-water interface, or the rates Of 2 consumption; nitrate consumption, and N2 production in the sediments, were calculated by using the equation given above from the averages of the concentrations during the steady period (Table 2). DISCUSSION Various methods have been developed to estimate bacterial denitrification in sediments. Billen (2) estimated denitrification simply from changes in nitrate concentrations during incubation of sediment samples. Koike and Hattori (6) and Oren and Blackburn (8) incubated a slurry of sediment samples together with [15N]nitrate and determined the 15N content in the dissolved N2 gas. Some investigators have used the acetylene inhibition technique of Yoshinari et al. (12) in work with coastal sediments. Srensen (1) injected acetylene-saturated distilled water into sediment core samples and followed the production of N2. Chan and Knowles (3) set a Plexiglas chamber on sediments and circulated water containing acetylene through the chamber. Other investigators have determined denitrification rates by measuring the production of N2 gas; Seitinger et al. (9) measured N2 production by undisturbed sediment cores incubated in a gastight chamber, and Kaplan et al. (4) measured N2 production by setting a bell jar on sediments. The method which we used in this study has the following advantages: (i) disturbance of sediment structure and associated chemical and physical changes are minimied, (ii) denitrification rates over a unit area of the sediments can be obtained directly, and (iii) rates of oxygen consumption and nitrate consumption can be determined simultaneously. The denitrification rates in Odawa Bay and Tokyo Bay (Table 2) fell within the range reported for sediments from other locations (Randers Fjord, 4.1 ng-atoms of N cm-2 h-1 [1]; Kysing Fjord,.7 ng-atom of N cm2 h-1 [8]). The Tama Estuary values were much higher. This probably resulted from eutrophication caused by abundant input of land drainage. High in situ concentrations of nitrate and nitrite in the overlying waters also might have been the cause of Downloaded from on January 3, 219 by guest

5 652 NISHIO, KOIKE, AND HATTORI APPL. ENVIRON. MICROBIOL. ODAWA BAY 15 MAY ' V Z2.' 2.! o 1 a 1! 1.1.!.5 -I IU A 1 6 o 5 o o.5 I TAMA ESTUARY 19 JUNE'8 2O, 1 o 2 -o 1 2 TIME (hours) the high ratio of nitrate consumption rates to oxygen consumption rates at this location (Table 2). Nitrate respiration might play an important role in the decomposition of organic matter in the Tama Estuary sediments. The contribution of nitrate respiration in the other locations seems minor. As a net flux, nitrate diffuses from the overly- TABLE 2. 3 FIG. 5. Changes with time in concentrations of N2 produced from nitrate in the experimental sedimentwater system. Samples were collected from Tokyo Bay (Stations 2 and 8, 1 September 198), Odawa Bay (15 May 198), and Tama Estuary (19 June 198). The amounts of N2 produced were calculated from the 15N contents in the N2 in the effluents. 1. _ Os -A.6 OA 2.1. TOKYO BAY 1 SEPT 8 * Sta.2 * o * a St& ing water to the sediment. However, if nitrate and nitrite are produced actively by nitrification in sediments, they diffuse from the sediment to the overlying water. This might reduce the 15N content in the overlying water. In preliminary experiments with [15N]ammonium, we found that the fluxes of nitrate plus nitrite which were produced by nitrification in the sediments were Oxygen consumption, nitrate (plus nitrite) consumption, and denitrification in Tama estuary, Odawa Bay, and Tokyo Bay sedimentsa Location Sampling date 2 consumption N3- (+NO2-) consumption Denitrification (nm cm-2 h-1) (ng-atoms of N cm-2 h-1) (ng-atoms of N cm-2 h-1) Tama Estuary 7 May (12) 3.7 (27.6) 19 June (138) 65.4 (78.9) Odawa Bay 15 May (9.) 4.66 (3.92) 2 June NDb.79 (.16) Tokyo Bay Station 2 1 September (6.96) 4.69 (3.26) Station 8 1 September (3.15) 2.73 (1.63) a The concentration of [15N]nitrate added was 1,uM, except for the experiment performed on 7 May 198. The rates were calculated by the equation given in the text, using the averages of the 2, nitrate (plus nitrite), and "5N2 concentrations present under steady-state conditions. The numbers in parentheses represent in situ rates calculated from the nitrate concentrations in the experimental system and the in situ concentrations of nitrate in the overlying water, assuming a linear relationship between the rates and nitrate concentrations. b ND, Not determined. Downloaded from on January 3, 219 by guest

6 VOL. 43, ng-atoms of N cm-2 h-1 at Tama Estuary and 7.3 ng-atoms of N cm-2 h-1 at Odawa Bay in May These fluxes of nitrate are equivalent to additions of.4 ng-atoms of N h-1 (Tama Estuary) and.7 ng-atoms of N h-1 (Odawa Bay) into the overlying water of our experimental system (approximately.7% [Tama Estuary] and 6.%o [Odawa Bay] of the nitrate supply from the reservoir). Actually, the 15N content in nitrate in the effluent remained unchanged within ca. ±1%, when Tama Estuary sediments were used. Therefore, isotope dilution of the added [15N]nitrate by nitrification did not seriously influence our estimates of denitrification rates except for the Odawa Bay sediment experiment (2 June 198), where the nitrate concentration in the overlying water was very low. ACKNOWLEDGMENTS We are grateful to E. Matsumoto for the collection of sediment samples in Tokyo Bay and D. D. Swinbanks for kind review of the manuscript. This work was supported by grants 5325 and 5332 from the Ministry of Education, Culture and Science, Japan. LITERATURE CITED 1. Bendachneder, K., and R. J. Robinson A new spectrophotometric method for the determination of nitrate in sea water. J. Mar. Res. 11: Billen, G A budget of nitrogen recycling on North Sea sediments off the Belgian Coast. Estuarine Coastal Mar. Sci. 7: DENITRIFICATION IN SEDIMENTS Chan, Y., and R. Knowles Measurement of denitrification in two freshwater sediments by an in situ acetylene inhibition method. Appl. Environ. Microbiol. 37: Kaplan, W., I. Vallela, and J. M. Teal Denitrification in a salt marsh ecosystem. Limnol. Oceanogr. 24: Kolke, I., and A. Hattori Denitrification and ammonia formation in anaerobic coastal sediments. Appl. Environ. Microbiol. 35: Kolke, I., and A. Hattori Estimates of denitrification in sediments of the Bering Sea shelf. Deep-Sea Res. 26A: Kodke, I., A. Hattorl, and J. J. Goering Controlled ecosystem pollution experiment: effect of mercury on closed water columns. VI. Denitrification by marine bacteria. Mar. Sci. Commun. 4: a.Nishio, T., I. Kolke, and A. Hattori N2/Ar and denitrification in Tama estuary sediments. Geomicrobiol. J. 2: Oren, A., and T. H. Blackburn Estimation of sediment denitrification rates at in situ nitrate concentrations. Appl. Environ. Microbiol. 37: Seltnger, S., S. Nixon, M. E. Q. Pilson, and S. Burke Denitrification and N2 production in near-shore marine sediments. Geochim. Cosmochim. Acta 44: Srensen, J Denitrification rate in a marine sediment as measured by the acetylene inhibition technique. Appl. Environ. Microbiol. 36: Wood, E. D., F. A. Armstrong, and F. A. Richards Determination of nitrate in sea water by cadmium-copper reduction to nitrite. J. Mar. Biol. Assoc. U.K. 47: Yoshinari, T., R. Hynes, and R. Knowles Acetylene inhibition of nitrous oxide reduction and measurement of denitrification and nitrogen fixation in soil. Soil Biol. Biochem. 9: Downloaded from on January 3, 219 by guest